Personalized Alignment of Language Models with Human Value Profiles
cs.HC, cs.AI
Submitted: 2026-09-01
Updated: 2026-09-01
Code: https://github.com/UMichHCI/SchwartzValueAlignment
Terminology
Sources
- Discovering Bias in Latent Space: An Unsupervised Debiasing Approach
- Refusal in Language Models Is Mediated by a Single Direction
- A General Language Assistant as a Laboratory for Alignment
- Constitutional AI: Harmlessness from AI Feedback
- Steerable Chatbots: Exploring Personalization Control Interfaces via LLM Activation Steering
- DailyDilemmas: Revealing Value Preferences of LLMs with Quandaries of Daily Life
- Towards Measuring the Representation of Subjective Global Opinions in Language Models
- Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaboration
- EVALUESTEER: Measuring Reward Model Steerability Towards Values and Preferences
- In-Context Alignment: Chat with Vanilla Language Models Before Fine-Tuning
- Aligning AI With Shared Human Values
- LoRA: Low-Rank Adaptation of Large Language Models
- Aligning Language Models to User Opinions
- Can Language Models Reason about Individualistic Human Values and Preferences?
- AI-Augmented Surveys: Leveraging Large Language Models and Surveys for Opinion Prediction
- Style Vectors for Steering Generative Large Language Model
- CLASH: Evaluating Language Models on Judging High-Stakes Dilemmas from Multiple Perspectives
- Programming Refusal with Conditional Activation Steering
- CultureLLM: Incorporating Cultural Differences into Large Language Models
- The Unlocking Spell on Base LLMs: Rethinking Alignment via In-Context Learning
Related papers
- EmoPhone: A Multi-Wave Dataset for In-the-Wild Mobile and Wearable Affect Sensing
- Tasteply: Externalizing Stable and Situational Preference as Manipulable Spatial Objects
- Shielded Trajectory Sampling for Personalized Relational Safety in Conversational AI
- From Tasks to Turns: Measuring Human Labor Contributions in Human-Agent Joint Production
- Reconstruction and Reflection of Positive Experiences through Resurfacing Laughter-indexed Everyday Moments
- LLM-Assisted Reranking to Operationalize Nuanced Objectives in Recommender Systems